ScummVM IRC Log - 16 Dec 2017

[Back to Index]

[00:03] --> dafioram joined #scummvm.
[00:09] <dafioram> snover: the raspian build you provided works on raspian stretch. From a fresh image i only had to install libmpeg and libsdl2. By chance was this a nonoptimized build?
[00:11] <-- dhewg left irc: Ping timeout: 260 seconds
[00:20] <dafioram> Even though it works i would not recommend playing scummvm on raspian stretch while in desktop mode as scummvm is very laggy and it maxes out the cpus. I imagine running with cmd line only will do better.
[00:30] Nick change: Deledrius_ -> Deledrius
[00:41] <-- dafioram left irc: Quit: Bye
[00:58] <tsoliman> I have an unreasonable urge to rebuild the toolchain from source (in stretch) instead of the jessie transplant
[01:29] <snover> aisdfhoiasdjc. i suck at life today
[01:29] <snover> why is travis sending bots into there to tell us that someones fork is failing&
[01:30] <-- TMM left irc: Remote host closed the connection
[02:02] <snover> i asked libretro to please fix that.
[02:06] --> DominusExult joined #scummvm.
[02:10] <-- Dominus left irc: Ping timeout: 260 seconds
[02:10] Nick change: DominusExult -> Dominus
[02:12] --> dhewg joined #scummvm.
[02:12] #scummvm: mode change '+o dhewg' by ChanServ!ChanServ@services.
[03:21] --> GitHub171 joined #scummvm.
[03:21] <GitHub171> [scummvm] dreammaster pushed 1 new commit to master: https://git.io/vborD
[03:21] <GitHub171> scummvm/master c969669 Paul Gilbert: XEEN: Implementing Arena location logic
[03:21] GitHub171 (GitHub171@gateway/service/github.com/x-bdemccpgmoeibapn) left #scummvm.
[03:31] <-- SylvainTV left irc: Read error: Connection reset by peer
[03:50] <-- dreammaster left irc:
[04:03] --> macdude22 joined #scummvm.
[04:07] <-- tsoliman left irc: Ping timeout: 240 seconds
[04:08] --> tsoliman joined #scummvm.
[04:08] #scummvm: mode change '+o tsoliman' by ChanServ!ChanServ@services.
[04:08] <-- balrog left irc: Ping timeout: 240 seconds
[04:08] <-- |Cable| left irc: Ping timeout: 240 seconds
[04:09] --> balrog joined #scummvm.
[04:13] <snover> tsoliman: youll be doing good by more people using your time on android instead
[04:15] <snover> _sev: thank you for your work on the disks, that sounds great. it should solve that problem for a long while, which will be really nice.
[04:18] <snover> lets see, who else did i fail to acknowledge in my stupor today&
[04:21] --> |Cable| joined #scummvm.
[04:26] <tsoliman> snover: I know .. that's what my gut is telling me too - it all came to a grinding halt anyway as one of the packages is missing a tiny original tarball
[04:27] <tsoliman> s/gut/brain/
[04:29] <tsoliman> I emailed toojays about it and gave up
[04:34] <-- ccawley2011 left irc: Quit: Page closed
[04:39] <snover> i tried to update a second compiler today and i regret it
[04:39] <snover> so take it from me
[04:40] <tsoliman> thanks! :) I needed that extra push
[04:40] <tsoliman> besides .. missing source = no go anyway
[04:40] <tsoliman> that's the story I am sticking with ;)
[05:08] <tsoliman> I am trying to build android again - configure says "Checking for OpenGL... no"
[05:09] <tsoliman> I am running `/data/sharedrepo/configure --host=android-v7a` which is what it seems the buildbot config does
[05:09] <tsoliman> according to workers/android/buildbot.cfg
[05:26] <snover> what makes you think it uses detected opengl?
[05:29] <snover> in configure it just unconditionally sets a linker flag instead of actually going through the proper detection process
[05:29] <tsoliman> that's what I suspected - this concept is new to me
[05:37] <tsoliman> the android cross compiler is slooooow - even compared to the maemo one
[05:39] <snover> that doesnt sound right
[05:40] <tsoliman> how long does a non-ccached build take for you (no configure options besides host)
[05:40] <tsoliman> over 5 mins?
[05:40] <snover> yes, a full build with any compiler takes like 15 minutes
[05:41] <tsoliman> I'm doing make -j4 ... (which is nprocs in that container) so it should be < 5 mins
[05:44] <tsoliman> I'll time it after it is done - I am sure I screwed something up
[05:47] <snover> what is your host os?
[05:48] <tsoliman> macOS 10.12.6
[05:48] <snover> did you use a bind mount to mount the source code?
[05:48] <tsoliman> yes
[05:48] <snover> https://blog.docker.com/2017/05/user-guided-caching-in-docker-for-mac/
[05:48] <tsoliman> btw - I just did a git pull .. .. git rudely plowed over my override yml
[05:49] <tsoliman> I dunno why I expected a conflict or a .bak or something
[05:49] <snover> it does give a conflict if there is a conflict unless you force it
[05:51] <tsoliman> I literally just did "git pull" .. it said Updating 73c14e5..a6c2d16 fast forward
[05:51] <snover> anyway, that blog post is a longwinded way of saying bind mounts are slow on macos unless you turn on caching https://docs.docker.com/compose/compose-file/#caching-options-for-volume-mounts-docker-for-mac
[05:52] <tsoliman> I think maybe because the file was .gitignored and then also added in the same commit
[05:53] <snover> i suppose that could do it. normally it wont pave over a file in the working directory though i dont know what happens when it introduces an ignored file. perhaps what you experienced!
[05:53] <tsoliman> ugh yeah it is a stupid git behavior .. I reproduced by doing a git reset --hard 73c14e5 .. then creating an override file .. and then doing a git pull .. and it again plowed over it
[05:53] <tsoliman> yeah I've had many many incidents of "oh no I can't pull .. you already have a file there"
[05:53] <tsoliman> and you'd have to be like "-f"
[05:54] <tsoliman> well what is gonna happen now is that my modifications to the override are gonna be marked "dirty"
[05:59] --> _sev_ joined #scummvm.
[05:59] #scummvm: mode change '+o _sev_' by ChanServ!ChanServ@services.
[05:59] <-- _sev left irc: Ping timeout: 264 seconds
[06:01] <snover> which is fine. you can always create your own separate overrides and specify with -f. i wish the use of the default overrides file was not needed, until docker-compose lets you override lists of things it will need to be this way for the production deployment to work correctly
[06:02] <snover> anyway, good luck, i need to head off for the evening. hopefully once you set up caching on the bind you will experience less slowness.
[06:03] <-- _sev_ left irc: Client Quit
[06:05] <tsoliman> thanks
[06:10] <tsoliman> snover: when you have time, can you shed some light on this https://pastebin.com/ZUCUazAT .. this is building your master with the buildbot worker "as standalone" .. configure --host=android-v7a and using CXX="ccache arm-linux-androideabi-clang++" as an env var for configure .. then "make" or "make androiddistdebug" both result in this linker error (latest docker image from dockerhub)
[06:10] <tsoliman> I'm guessing you don't get that because only SCI is enabled
[06:13] <tsoliman> this whole thing is dejavu - I might be up too late
[06:20] <logix> tsoliman: not sure if this helps much, I'm unsure about atof but srand, rand and stdin are mentioned on https://github.com/openssl/openssl/issues/3826
[06:20] <logix> "When creating Android NDK toolchain using make_standalone_toolchain.py, pass --deprecated-headers as an argument. Found it reading NDK's changelog."
[06:20] <logix> "Note: The deprecated headers will be removed in a future release, most likely r16. If they do not work for you, file bugs now."
[07:51] <tsoliman> snover: also I noticed an inconsistency (which might be intentional) between the 2 CXX vars .. the docker image says arm-linux-androideabi-c++ but buildbot says ccache arm-linux-androideabi-clang++ .. (c++ vs clang++) .. clang++ is a shell script (it's not a symlink to the same thing)
[07:55] <tsoliman> and this is where the problem is I think .. using c++ causes all engines to be enabled .. but using clang++ causes Sword25 and wintermute (and a few others) to be disabled because of missing dependencies (JPG, PNG, Vorbis, etc)
[07:56] <tsoliman> so no I don't think the inconsistency is intentional
[08:05] <-- Asterisk left irc: Ping timeout: 264 seconds
[08:08] --> Asterisk_ joined #scummvm.
[08:11] <-- Strangerke left irc: Ping timeout: 264 seconds
[08:19] Nick change: Asterisk_ -> Asterisk
[08:19] <-- Asterisk left irc: Changing host
[08:19] --> Asterisk joined #scummvm.
[08:22] <-- LittleToonCat left irc: Remote host closed the connection
[09:03] --> Henke37 joined #scummvm.
[09:48] --> Strangerke joined #scummvm.
[10:12] --> criezy joined #scummvm.
[10:12] #scummvm: mode change '+o criezy' by ChanServ!ChanServ@services.
[10:31] --> ajax16384 joined #scummvm.
[10:31] #scummvm: mode change '+o ajax16384' by ChanServ!ChanServ@services.
[11:12] <-- TAS_2012v left irc: Read error: Connection reset by peer
[11:13] --> TAS_2012v joined #scummvm.
[11:24] --> GitHub170 joined #scummvm.
[11:24] <GitHub170> [scummvm-web] criezy pushed 2 new commits to 2.0: https://git.io/vbo5k
[11:24] <GitHub170> scummvm-web/2.0 2365209 Thierry Crozat: WEB: Add icon for RISC OS platform
[11:24] <GitHub170> scummvm-web/2.0 0846e05 Thierry Crozat: WEB: Add Haiku and RISC OS packages to download page
[11:24] GitHub170 (GitHub170@gateway/service/github.com/x-ypellewbkklwsfwj) left #scummvm.
[11:27] <criezy> rootfather: I have added ReleaseNotes_DE to frs
[11:59] --> TMM joined #scummvm.
[11:59] <-- TMM left irc: Changing host
[11:59] --> TMM joined #scummvm.
[11:59] #scummvm: mode change '+o TMM' by ChanServ!ChanServ@services.
[12:32] <-- Lightkey left irc: Ping timeout: 265 seconds
[12:44] --> Lightkey joined #scummvm.
[13:18] <-- Henke37 left irc: Quit: ERR_SHUTDOWN
[13:21] <-- rootfather left irc: Ping timeout: 248 seconds
[13:22] --> rootfather joined #scummvm.
[14:00] <-- antlarr left irc: Ping timeout: 256 seconds
[14:00] --> antlarr2 joined #scummvm.
[14:28] <-- TMM left irc: Ping timeout: 248 seconds
[14:57] --> Farmboy0 joined #scummvm.
[14:57] <-- Farmboy0 left irc: Changing host
[14:57] --> Farmboy0 joined #scummvm.
[14:57] --> TMM joined #scummvm.
[14:57] <-- TMM left irc: Changing host
[14:57] --> TMM joined #scummvm.
[14:57] #scummvm: mode change '+o TMM' by ChanServ!ChanServ@services.
[15:00] <-- ajax16384 left irc: Quit: Leaving
[15:00] --> dreammaster joined #scummvm.
[15:00] #scummvm: mode change '+o dreammaster' by ChanServ!ChanServ@services.
[15:04] --> GitHub123 joined #scummvm.
[15:04] <GitHub123> [scummvm] dreammaster pushed 1 new commit to master: https://git.io/vbojU
[15:04] <GitHub123> scummvm/master 93b9eb4 Paul Gilbert: XEEN: Added remainder of Arena location
[15:04] GitHub123 (GitHub123@gateway/service/github.com/x-dyrnvzltuoefvjhy) left #scummvm.
[15:32] --> GitHub163 joined #scummvm.
[15:32] <GitHub163> [scummvm] dreammaster pushed 1 new commit to master: https://git.io/vbKvT
[15:32] <GitHub163> scummvm/master 808dd29 Paul Gilbert: XEEN: Convert character _awards array from bool to int...
[15:32] GitHub163 (GitHub163@gateway/service/github.com/x-yjrycbdvcmgrhwjx) left #scummvm.
[15:58] --> Henke37 joined #scummvm.
[16:07] --> dafioram joined #scummvm.
[16:17] <dafioram> snover: I would like to make the case for building the raspberry pi release with libpng12 instead of libpng16. The reason is because it will allow playing scummvm on other raspberry pi OSs (besides stretch) easier. While stretch has libpng16 available through apt-get Jessy only has libpng12 from apt-get. Obviously, a user could build 16.34 from source on jessy to get libpng16, but I don't think thats something we should force. I
[16:20] <snover> there is no libpng12 in stretch, only libpng16.
[16:20] --> ccawley2011 joined #scummvm.
[16:20] <snover> also your message was cut off at force. I
[16:22] <dafioram> this link shows packages available on jessy and stretch. It says stretch libpng12-dev is in stretch and jessy. http://www.raspberryconnect.com/raspbian-packages-list/item/109-raspbian-libdevel
[16:23] --> GitHub174 joined #scummvm.
[16:23] <GitHub174> [scummvm] dreammaster pushed 1 new commit to master: https://git.io/vbKJi
[16:23] <GitHub174> scummvm/master d2dbf59 Paul Gilbert: XEEN: Remove old deprecated method declaratations
[16:23] GitHub174 (GitHub174@gateway/service/github.com/x-cyqdblnheeisyweo) left #scummvm.
[16:23] <dafioram> but by default libpng-dev gives u libpng16 (on stretch)
[16:24] <dafioram> I think there is more jessys in the wild then stretchs since raspian stretch is only 4 months old
[16:26] <dafioram> also my libpng12 compile scummvm plays on stretch so either the api is compatible or stretch includes libpng12 as well as libpng16
[16:35] <dafioram> stretch doesnt have libpng12 by default so I guess the api are backwords compatible
[16:45] <dafioram> if it easys you mind we could have two raspberry versions one for stretch and one for jessy
[16:46] <snover> ok, there is no libpng12 in the official debian repository so i guess whoever is in charge of that different repository backported it.
[16:46] <snover> there are better ways to handle this, like just bundling the dependencies or statically linking.
[16:46] <snover> in terms of creating multiple builds, thats the least preferred thing to do
[16:49] --> SylvainTV joined #scummvm.
[16:49] #scummvm: mode change '+o SylvainTV' by ChanServ!ChanServ@services.
[16:57] <dafioram> snover: lets say we build with libpng12 dynamically and i test it on raspian stretch with libpng16 only installed and i am able to take screenshots and view thumbnails successful. If that went well could their be other uintented issues?
[16:58] --> Littleboy joined #scummvm.
[16:58] #scummvm: mode change '+o Littleboy' by ChanServ!ChanServ@services.
[16:59] --> GitHub121 joined #scummvm.
[16:59] <GitHub121> [scummvm] dreammaster pushed 1 new commit to master: https://git.io/vbKTM
[16:59] <GitHub121> scummvm/master d0d2792 Paul Gilbert: XEEN: Show Party dialog when signing into taverns
[16:59] GitHub121 (GitHub121@gateway/service/github.com/x-wxkguykuavurjxqf) left #scummvm.
[16:59] <dafioram> I guess statically build in libpng16 would be less error prone and allows the latest version to be used?
[17:02] <snover> tsoliman: huh, thats weird. maybe check the config.log to see what is happening, otherwise i will take a look in a bit since i need to do some updates and then start doing the full build tests
[17:11] <-- dafioram left irc: Quit: Bye
[17:12] <-- dreammaster left irc:
[17:24] <-- mortalius_ left irc: Quit: WeeChat 1.9
[17:24] --> mortalius joined #scummvm.
[17:34] --> LittleToonCat joined #scummvm.
[17:57] --> GitHub117 joined #scummvm.
[17:57] <GitHub117> [scummvm-web] criezy pushed 1 new commit to 2.0: https://git.io/vbKIh
[17:57] <GitHub117> scummvm-web/2.0 2e1a215 Thierry Crozat: WEB: Some more work on the 2.0.0 release announcement
[17:57] GitHub117 (GitHub117@gateway/service/github.com/x-mxovkwzlsvkeujss) left #scummvm.
[18:01] <snover> criezy: thank you! :
[18:01] <snover> :)
[18:11] <wjp> hurray, I made the "extremely good looking" category
[18:20] --> ny00123 joined #scummvm.
[18:31] --> GitHub187 joined #scummvm.
[18:31] <GitHub187> [scummvm-web] csnover pushed 1 new commit to 2.0: https://git.io/vbKtS
[18:31] <GitHub187> scummvm-web/2.0 352e0c2 Colin Snover: WEB: Minor fixes to 2.0 release news...
[18:31] GitHub187 (GitHub187@gateway/service/github.com/x-xitkwqdclppnoiql) left #scummvm.
[18:55] --> Strangerke_ joined #scummvm.
[18:58] <-- Strangerke left irc: Ping timeout: 272 seconds
[18:58] Nick change: Strangerke_ -> Strangerke
[19:18] --> _sev_ joined #scummvm.
[19:18] #scummvm: mode change '+o _sev_' by ChanServ!ChanServ@services.
[19:22] --> ajax16384 joined #scummvm.
[19:22] #scummvm: mode change '+o ajax16384' by ChanServ!ChanServ@services.
[21:21] <-- _sev_ left irc: Ping timeout: 248 seconds
[21:29] <-- ajax16384 left irc: Read error: Connection reset by peer

[21:50] --> GitHub39 joined #scummvm.
[21:50] <GitHub39> [scummvm-web] criezy pushed 2 new commits to 2.0: https://git.io/vbKW4
[21:50] <GitHub39> scummvm-web/2.0 078f515 Thierry Crozat: WEB: Clean list of contributors for the 2.0.0 release announcement
[21:50] <GitHub39> scummvm-web/2.0 f2392f9 Thierry Crozat: WEB: Translate 2.0 release news to French
[21:50] GitHub39 (GitHub39@gateway/service/github.com/x-jtczchqnadxcahyl) left #scummvm.
[21:50] --> Joefish joined #scummvm.
[21:50] #scummvm: mode change '+v Joefish' by ChanServ!ChanServ@services.
[21:50] <snover> criezy: i should have a tools build for you as soon as i can actually get wxwidgets to compile successfully.
[21:51] <criezy> Great.
[21:52] <snover> its Makefile is almost 2MB. i have no idea why they have a configuration switch for like every UI control
[21:54] <snover> of *course* i must have only a radio button and not a check box
[21:55] <snover> F!@#$!@#$HJ.
[21:55] <snover> and their dependency management is non-existent
[21:56] <snover> try to disable building xml and get ready for modules like richtextbuffer to randomly fail to compile because some XML types arent defined
[21:57] <criezy> I think I basically just built it with the default, although I can't test right now as it is on my other computer, and I don't really remember as it was years ago.
[21:57] <criezy> *I can't check right now
[21:58] <-- ny00123 left irc: Quit: Leaving
[21:58] <snover> try to enable XML and get ready for its built in expat to fail to compile because reasons
[21:58] <snover> No rule to make target 'src/expat/lib/xmlparse.c'
[22:07] <snover> with external expat, hopefully this will work now.
[22:08] --> dreammaster joined #scummvm.
[22:08] #scummvm: mode change '+o dreammaster' by ChanServ!ChanServ@services.
[22:10] <snover> i did not intend to procrastinate until the last possible moment on this, for some reason i had the 20th in my head
[22:25] --> GitHub191 joined #scummvm.
[22:25] <GitHub191> [scummvm-tools] criezy pushed 6 new commits to master: https://git.io/vbK8l
[22:25] <GitHub191> scummvm-tools/master ad0806d Willem Jan Palenstijn: COMMON: Rename String::printf to format to match ScummVM
[22:25] <GitHub191> scummvm-tools/master 9acf70e Willem Jan Palenstijn: COMMON: Add some Common::String/std::string interoperability functions
[22:25] <GitHub191> scummvm-tools/master 4f05f4e Willem Jan Palenstijn: PRINCE: Replace C++11 string functions
[22:25] GitHub191 (GitHub191@gateway/service/github.com/x-ulvaprllbvwdeysx) left #scummvm.
[22:50] --> GitHub171 joined #scummvm.
[22:50] <GitHub171> [scummvm] dreammaster pushed 1 new commit to master: https://git.io/vbK49
[22:50] <GitHub171> scummvm/master 8c850f8 Paul Gilbert: TSAGE: R2R: Clicking F5 fro Options dialog
[22:50] GitHub171 (GitHub171@gateway/service/github.com/x-qeibwowcnlmpzvzx) left #scummvm.
[23:32] --> GitHub20 joined #scummvm.
[23:32] <GitHub20> [scummvm] dreammaster pushed 1 new commit to branch-2-0: https://git.io/vbK0z
[23:32] <GitHub20> scummvm/branch-2-0 16a6089 Paul Gilbert: TSAGE: R2R: Clicking F5 fro Options dialog...
[23:32] GitHub20 (GitHub20@gateway/service/github.com/x-jhwowveuoujcykwc) left #scummvm.
[23:34] --> BeefEats joined #scummvm.
[23:36] <-- BeefEats left irc: Client Quit
[00:00] --- Sun Dec 17 2017